Shear Capacity Prediction of FRP Reinforced Concrete Beams Using Hybrid GMDH–GA

Abstract

In recent years, the use of composite rebars in reinforced concrete structures has received much attention due to its high corrosion resistance, significant tensile strength, and appropriate non-magnetization characteristics. Due to the lower modulus of elasticity of composite rebars than steel rebars, concrete beams reinforced with composite rebars have relatively lower shear strength compared to beams reinforced with steel rebars. On the other hand, shear failure in concrete beams reinforced with composite rebars is generally brittle and requires accurate prediction of the behavior of these members. Therefore, in this study, the shear strength of concrete beams reinforced with composite rebars is predicted using a combination of GMDH type neural networks and genetic algorithms based on a wide range of experimental results. The key effective parameters that consider in this study are the width of the web, effective depth of the beam, shear span to depth ratio, concrete compressive strength, modulus of elasticity of FRP longitudinal bars, and longitudinal reinforcement ratio. The accuracy of the proposed method has been verified by comparing the model predictions with the collected experimental results and existing shear design equations. The results show that the proposed model has more accurate results in calculating the shear strength of concrete beams than other existing relationships. A sensitivity analysis is also performed to assess the effect of the input parameters on the shear strength of FRP-reinforced concrete beams.